The National Health Interview Survey (NHIS) is one of the longest-running federal surveys and is the principal source for information on the changing health of Americans. Presently, however, the NHIS is underutilized for the study of long-term change. The complexities of the survey design and file structures of the existing public use files make the NHIS unwieldy even for analysis of single years; the incompatibilities across survey years--and particularly across survey design periods--make the cost of long-run chronological analysis prohibitive for many researchers. This proposal seeks funding to create an Integrated Health Interview Series (IHIS) comprised of NHIS data and documentation for the period 1963 to 2003. The IHIS will simplify access to approximately 6,000 variables on over three million respondents contained in over 300 different data files. It will multiply the value of NHIS data by allowing researchers to make consistent comparisons through four decades of dramatic change in public health, and thus to study the health status of Americans as a dynamic process. The IHIS will result in a substantial body of new scientific and policy-relevant research into health behavior and disparities, access to and use of medical care, population aging, progress toward public health goals like the Health People 2010 initiatives, and many other topics. The project leverages previous federal investments in collecting the NHIS data and in data infrastructure projects at the Minnesota Population Center. The IHIS project consists of four interrelated components: 1) a design component to develop a comprehensive data integration plan covering data structures, variable coding schemes, and new variables to enhance usability; 2) a data improvement component to create new public use files for 1963 to 1968 and to allocate missing and inconsistent data values throughout the data series; 3) a documentation component offering comprehensive integrated XML metadata, including extensive coverage of the comparability of each variable across time; and 4) a data access component allowing web-based retrieval of data and documentation, with capabilities for variable search, automated file merging, and data reformatting.